Advances on Deep Learning Algorithms in the context of Feature Selection on Collision Detection in Internet of Vehicles: Trends, Challenges, and Future Prospects

Abstract

The Internet of Vehicles (IoV) heralds a transformative era in intelligent transportation, enabling real-time data exchange among vehicles, infrastructure, and sensors. However, the prevalence of vehicular collisions remains a critical challenge requiring intelligent, data-driven solutions. This article explores recent advances in deep learning (DL) techniques for collision detection within the IoV landscape. It systematically discusses the integration of DL models such as CNNs, LSTMs, GANs, and NSGA-III in addressing collision prediction, object recognition, and feature selection. The study also examines hybrid DL models, sensor fusion approaches, and their applications in vehicular safety and infrastructure-aware decision systems.